Former Harvard Business School dean says new CEOs keep making the same mistakes, starting with trying to run the company

Nitin Nohria, who ran Harvard Business School from 2010 to 2020, says newly minted chief executives routinely misjudge what the job actually requires, and some don't find out until they're fired.

Nohria laid out the most common blind spots on the HBR IdeaCast in an episode released Tuesday, ahead of his new book, The CEO: The Role, the Reality, the Responsibility , which Harvard Business Review Press is publishing on September 29, 2026. amazon

The book draws on a quarter century of twice-yearly HBS New CEO Workshops that thousands have attended. Here's what Nohria says new CEOs get wrong. google

Mistake No. 1: Trying to run the company

Most executives earn the top job by running a business unit or function exceptionally well. That's exactly the habit they need to drop, Nohria said.

A CEO's job is to set up the people who report to them to run the company, he said. When the CEO wades into operations, the whole organization starts checking every decision with the top, and work grinds to a halt.

The CEO's voice works like a megaphone, Nohria said. Even a small intervention signals to everyone below that they need sign-off.

Mistake No. 2: Believing the information that reaches you

New CEOs assume their access to information will improve. Technically it does; they can ask anyone anything.

The catch, Nohria said, is that everyone reporting to the CEO has an agenda, even if it's just not wanting to bother the boss with a problem they hope to fix by Friday.

He passed along a metaphor from a colleague in the cereal business: information starts out as plain corn flakes and arrives on the CEO's desk frosted.

His fix is simple, though CEOs rarely do it. Go talk to frontline employees and customers. Nohria said time-use studies show those are the people who get the least time on a CEO's calendar, even though they're the most likely to tell the truth.

Mistake No. 3: Misreading the board

New CEOs are good at managing one boss. A board is 10 to 12 bosses who meet every few months, each with a vote, and who tend to raise concerns gently.

That politeness is dangerous, Nohria said. CEOs who lose touch with individual directors mistake courtesy for support.

"I have yet to meet a CEO who got fired who didn't feel surprised," he said.

The remedy: get to know each director privately and draw out the quiet ones, who Nohria said are sometimes more influential than the loudest voices in the room.

The red flags Nohria watches for

Nohria said he looks for a few warning signs that a CEO is headed for trouble:

• No clear agenda by year two. He said he sometimes asks five randomly chosen senior leaders to describe the CEO's priorities. If they can't, the CEO is unlikely to succeed.

• Blaming the analysts. A CEO who keeps insisting Wall Street doesn't understand the company is usually ignoring investors, he said.

• A losing streak. Boards give some slack, Nohria said, but four or five disappointing quarters in a row puts a CEO on thin ice.

The trait he looks for most in a new CEO is humility: projecting confidence while staying open to learning the job.

Audit your calendar every quarter

Nohria's most concrete piece of advice: write down your personal agenda at the start of each quarter, then check your calendar at the end.

If less than about half your time went to those priorities, you're wasting your scarcest resource, he said. CEOs can hire, fire, and move money. They can't make more hours.

He also said new CEOs should evaluate an inherited leadership team as if building it from scratch, and act fast. The most damaging pattern he's seen is postponing changes a CEO already knows need to happen.

AI changes the 20%, not the 80%

Nohria estimated that roughly 80% of the CEO job has stayed constant over three decades. The other 20% shifts with the times, from social-issue stances a few years ago to tariffs, war, inflation, and AI today.

On AI, he urged discipline. Pick the two or so initiatives that fit your company, he said, and have the courage to say you won't chase the rest.

Boards are slowing down on CEO swaps

The book lands as corporate America cools off after a period of heavy leadership churn. Through November 2025, 419 CEOs had exited publicly traded companies, the most since Challenger began tracking company type in 2010. challengergray

This year looks different. Challenger reports CEO exits are down 26% year-to-date as companies prioritize leadership stability. challengergray

That makes the first few years of a new CEO's tenure more consequential. Nohria said most CEOs climb a steep learning curve and feel they understand the job by year two or three.

His advice for anyone still aiming for the corner office: crush your current job, and raise your hand for the risky assignment nobody else wants.

Nohria's observations come from his workshops and research; the interview did not detail the data behind specific benchmarks, such as the half-of-your-time calendar rule.